Constraints on percussive seismic signals in a noisy environment by European fiddler crabs, <i>Afruca tangeri</i>

Many animals communicate using seismic vibrations. Signaller morphology, signal production method and environmental factors impose interacting constraints that may be impossible to replicate in the laboratory, making it essential to study seismic communication in situ. Here, we focused on the constraints on percussive seismic signals in European fiddler crabs (Afruca tangeri), recording a large dataset of percussive seismic signals in situ, and testing for waveform differences as a function of signaller morphology and behaviour. In addition, we aimed to characterise signal degradation and interference by seismic noise from wind and vibrated vegetation in the natural environment. We obtained over 8000 percussive seismic signal recordings, and found that although the length, rhythm and loudness of the signals all varied as a function of behaviour, their frequency content did not. Consequently, behaviours could be discriminated based on seismic recordings alone. Larger claws were only associated with louder signals in the case of claw drumming behaviours, but morphology did not affect percussive signal features otherwise. Environmental effects on percussive signals were substantial as signals attenuated significantly over distance, and wind speed was positively correlated with seismic noise, albeit independently of distance to vegetation. We conclude that percussive seismic signals are limited in their ability to convey information through frequency, but that their broadband nature is advantageous in the face of noise and frequency filtering by the substrate. In contrast, changing the amplitude and repetition rate of a percussive signal offers a simple but effective means for small animals to communicate seismically in noisy environments.